Common control units today, in particular control units in motor vehicles, pass through an initialization phase after power-on. During this initialization phase, for example, control programs are loaded to the control unit's main memory, values previously stored in a memory area are loaded to the main memory, values are input from sensors that are connected to the control unit via data lines, and/or the input values are subjected to plausibility checks.
An initialization phase of this type lasts hundreds of milliseconds. If this initialization phase begins with a start request, for example by turning the ignition key, the actual start of the internal combustion engine is delayed by this period of time.
In modern internal combustion engines, the initialization of a control unit also involves synchronizing the internal combustion engine with the control unit. This is necessary because information about a present state of the internal combustion engine must be available to start the internal combustion engine. For example, to correctly control the injection and ignition apparatus, the position of at least one cylinder must be detected and transmitted to the control unit. Once the position of a cylinder is known, the control unit is able to determine the positions of the remaining cylinders therefrom. In the case of a cylinder 1 detection, the position of a cylinder identified as “cylinder 1” is detected for this purpose.
A control unit is typically initialized by a request from the user to start the internal combustion engine (start request). The required synchronization of the control unit with the internal combustion engine is carried out, for example, while the internal combustion engine is being placed in rotary motion by the starter. Suitable sensors are used to detect cylinder 1, which may last as long as two crankshaft rotations. Fuel is then injected, and the resulting fuel-air mixture is ignited in the combustion chamber of a cylinder. A period of over one second may therefore pass from the start request to actual starting of the internal combustion engine, which is a nuisance to the user.
A method for activating interconnected network components that are preferably installed in a vehicle is described in German Patent Application No. DE 198 53 451. One network component of this type, for example, is a control unit that communicates with measuring devices via a network, for example a bus system such as the Controller Area Network (CAN). A detecting device generates a signal when a request to start the network components is likely (start request probability). As a function of a signal of this type, a first network component sends a message via the bus system, which activates the remaining network components. The first network component must be either permanently active or have a signal input via which this first network component is activatable when a signal is present at this input. If a start request probability is detected, all network components are activated. However, if no use is actually made thereof because a user merely opened the vehicle door to remove an item located in the vehicle, for example, the active network components are deactivated again after a predetermined interval. The network components are reactivated if a usage probability is redetected. As a result, the entire network may be activated and deactivated multiple times without an actual starting operation taking place. This consumes power unnecessarily. In particular, this method does not result in the control unit being synchronized with the internal combustion engine.